key: cord-0844889-1pfr7td0
authors: Bilò, M.Beatrice; Braschi, M.Chiara; Piga, Mario A.; Antonicelli, Leonardo; Martini, Matteo
title: Safety and adherence to venom immunotherapy during COVID-19 pandemic
date: 2020-11-26
journal: J Allergy Clin Immunol Pract
DOI: 10.1016/j.jaip.2020.11.030
sha: 1760b22a3d83ec8987fdb6a17cb5e0630d0e348b
doc_id: 844889
cord_uid: 1pfr7td0

Background According to expert consensus, the time interval between Hymenoptera venom immunotherapy (VIT) injections can be extended up to 12 weeks, without significant impact on efficacy and safety. However, the COVID-19 pandemic caused longer delays, and no recommendations are available to manage this huge extension. Objectives Our study aims to provide advice on how to resume VIT safely after a long delay from the last injection considering the potential risk factors for side effects (SE), without starting again with the induction phase. Methods All the patients who delayed VIT due to pandemic were consecutively enrolled in this single-centre study. The time extension was decided according to their risk profile (e.g. long pre-pandemic time interval, severe pre-VIT reaction, older age, multi-treatments), and correlation analyses were performed to find potential risk factors of SE. Results The mean delay from the pre- (7 weeks) to the post-pandemic VIT interval (15.5 weeks) was 8.5 weeks. The total amount of the pre-pandemic VIT maintenance dose was safely administered in one day in 78% of patients, while only three experienced SE, out of 87, and their potential risk factors were identified in bee venom allergy and recent VIT initiation. Conclusions In a real-world setting, long VIT delays may be safe and well-tolerated, but more caution should be paid in resuming VIT in patients with long pre-pandemic maintenance interval, severe pre-VIT reaction, recent VIT initiation, older age, multi-drug treatments, and bee venom allergy. This is useful in any case of long, unplanned, and unavoidable VIT delay.

No funding was received for this work. What is already known about this topic? 56 The COVID-19 pandemic raised the need to extend venom immunotherapy (VIT) time intervals safely. 57

However, no guidelines are currently available on how to manage long extensions of intervals between VIT 58 administrations (i.e. > 3 months). 59

What does this article add to our knowledge? 60 Our results show that long VIT delay may be tolerated in clinical practice, if key risk factors are considered: 61 bee venom, long time interval between injections, severe pre-VIT reaction, date of VIT initiation, age, multi-62 treatments. 63

How does this study impact current management guidelines? 64 A safe and effective protocol to resume VIT after a long interruption is useful in case of unexpected delays 65 in treatment. A new build-up phase may be safely avoided in most cases, with compliance and cost 66 days, and the ultra-rush protocol maintenance dose is reached within 1-2 days or within a few hours. 91

In Europe, VIT may be performed with aqueous (non-purified or purified) extracts and depot preparations 92 of Hymenoptera venoms, the last being used only for the conventional build-up and maintenance schedule. 93

Many European allergists switch to depot preparations following the up-dosing phase with an aqueous 94 extract. 1 95 According to expert consensus, injections are usually given every four weeks in the first year of treatment, 96 every six weeks in the second year, and in case of a five-year treatment every eight weeks from year 3-5. 2 If 97 lifelong treatment is necessary, extending the maintenance interval (MI) to three months may be relevant 98 in terms of convenience and economic savings, as it does not seem to reduce effectiveness or increase side 99 effects (SE). 2,3 100

During COVID-19 pandemic, allergists and immunologists continued to play their important role in the 101 prevention of venom anaphylaxis, 4 especially in the management of VIT, its delays in administration, and 102 the need to reduce the hospital admissions at the same time. Currently there are suggestions on how to 103 behave in the event of a pandemic, 4-7 without a detailed approach on how to resume VIT after a long delay 104 from the last injection, avoiding to start again VIT. 105

The aim of the present study is to share the experience of a single specialized Allergy Unit in Italy on this 106 topic, focusing on the key factors taken into account to safely extend the time intervals between VIT 107 injections limited to the period of the pandemic, and on the characteristics/risk factors of patients who 108 experienced SE because of that extension. 109 J o u r n a l P r e -p r o o f METHODS   110   This study consecutively enrolled, over a period of three months, all allergic patients treated with  111   maintenance VIT, who delayed their shots because of the COVID-19 pandemic.  112 Our Allergy Clinic collects patients coming from central and southern Italy and is specialized in performing 113 rush and ultra-rush VIT protocols. 114

Whenever discontinuation was not recommended or patients preferred to continue VIT because of quality 115 of life issue, the date of the next appointment and the number of VIT shots were decided on the basis of 116 several patient's features, like the pre-VIT severity reaction, diagnosed or potential mast cell disorders 117 (based on a positive REMA score), 8 skin testing, age, duration of VIT, type of HVA/venom extract, time 118 interval between VIT injections before the COVID-19 pandemic (pre-pandemic MI), comorbidities and 119 pharmacological treatments. 120

Before resuming VIT, all patients underwent a medical examination, and spirometry and/or ECG were 121 performed if necessary. A venous access was also placed, and the anesthesiologist was always alerted. 122

Patients were not allowed to attend the visit in case of quarantine or previous contacts with COVID-19 123 cases, as assessed by the phone contact before the visits. Each patient was screened for COVID-19 potential 124 infection/exposure with epidemiological interviews and body temperature check at entrance, as per 125 hospital's COVID-19 contingency plan, and provided with disinfectant and personal protective equipment. 126

During the visits, social distancing was respected whenever possible. 127 A post hoc analysis was performed to analyze the relationship between the delay in VIT administrations due 128 to the COVID-19 pandemic and the key baseline characteristics of patients considered for the risk 129 assessment of the VIT time interval prolongation (Pearson's correlation or Spearman's test, when 130 applicable, and ANOVA test for VIT duration). The pre-pandemic MI is defined as the original time interval 131 between VIT injections, before pandemic. The post-pandemic MI is the new time interval between VIT 132 injections, that was prolonged due to pandemic; it reflects the total magnitude of the VIT interruption (i.e. Out of 292 patients treated with maintenance VIT, 177 (61%) respected the original appointment despite 151 the COVID-19 pandemic, and 28 (9.6%) stopped the treatment either due to patient's own decision (6 152 subjects), or in agreement with the investigators, as eligible to VIT discontinuation 1,2 (22 subjects). Eighty-153 seven patients (30%) delayed their appointments because of fear of SARS-CoV-2-infection. The features of 154 these patients are shown in Table I . In particular, 72% of them were males, 84% have had a history of 155

Mueller class 3-4 and 4.5% were suffering from comorbidities (mainly represented by cardiovascular 156 diseases), 12% were treated with anti-hypertensive drugs. Four patients were suffering from diagnosed 157 mastcell disorders, while 29% of the patients showed a REMA score ≥ 2, and 10% STL ≥ 11.5 mcg/l. The 158 mean duration of VIT was 6.1 ± 5.5 years (15% on VIT for ≤ 1 year), while the mean pre-pandemic MI was 7 159 ± 2.3 weeks (min-max: 2-12 weeks). Seventy-four percent of patients were treated with vespid venoms, 160 and 10% with 2 vespid extracts, the remaining patients with bee venom; the majority of extracts used were 161 depot. The total maintenance dose was 100 mcg in all patients but three, in whom it was 150 mcg. None of 162 these patients was restung during the months of the pandemic. No cases of SARS-CoV-2 infection, 163 quarantine, or contact with potential infected people have been reported among our study population. 164 165 VIT during and post-pandemic: timing 166 The mean post-pandemic MI (i.e. total VIT interruption due to pandemic) was 15.5 ± 3.3 weeks; it was ≥ 4 167 months in 56% of patients, with the longest in 8% (≥ 5 months, max: 22 weeks). The mean delay (i.e. 168 extension from pre-to post-pandemic MI) was 8.5 ± 3.1 weeks, and it was ≥ 2 months in 59% of patients 169 (max: 16 weeks) (Table II) . VIT was administered in one day through 3 or 4 shots in all but one patient, who 170 received only 2 shots (Table III) . The same type of extract (aqueous or depot) from the pre-COVID pandemic 171 was reused. The total amount of the pre-pandemic maintenance dose of venom was administered in one 172 day in 78% of patients; the others received from 60% to 85% (15% of patients) or 45-50% (7% of patients) 173 of the total dose in the first day, and then the maintenance dose was reached at the subsequent visit (Table III ). The reasons for not reaching the total dose in one day, in these patients (22%), was mostly due to 175 logistic issues related to the pandemic re-organization. Only 3 patients (3%) did not manage to tolerate the 176 total dose in one single day due to SE. All the other patients did not show any SE, including subjects on VIT 177 for a short time (less than one year, but more than one month) and with long VIT interruptions (up to 5.5 178 months). 179

Once VIT was resumed after the long pandemic lapse, and the pre-pandemic dosage was reached, the MI of 180 subsequent shots was also kept mostly at the pre-pandemic MI. 181

The analysis shows a linear negative correlation between pre-pandemic MI and the delay in VIT 182 administration due to pandemic (Pearson's r= -0.272; p=0.011, 95% CI: -0.456 to -0.065) (Figure 1 ). Indeed, 183 the mean delay decreases by an estimated 0.363 weeks per one week increase in pre-pandemic MI 184 (p=0.011, 95% CI: -0.639 to -0.086). Therefore, the longer the pre-pandemic MI, the shorter was the delay 185 allowed in VIT administration (Table II) . There is a trend for a negative correlation between VIT duration and delay in VIT administration, but this 191 correlation is not statistically significant (p=0.085). This is supported by the fact that no significant 192 differences, in terms of delay, are observed across different years of VIT duration (F=0.69; p=0.603). 193

Patients undergoing concomitant medications had a shorter delay (p=0.041), as well as older patients 194 (p=0.017, Figure 2B ). Namely, the mean delay in ≥65-year-old patients was 7.1 ± 2.6 weeks, compared to 195 9.2 ± 3.1 weeks of the younger ones. 196

On the other hand, no significant correlation has been observed between VIT delay and gender, type of 197 HVA, comorbidities, pre-pandemic skin test reactivity, previous SE related to VIT According to literature, in case of a 5-year treatment 2 or of more than 4-year treatment 3 progressively 230 extending the MI to 3 months does not seem to increase the risk of a SR or reduce VIT effectiveness. A 6-231 month interval does not seem to influence VIT safety 9,10 , but it is less effective in the case of a few years of 232 bee VIT 9 . Moreover, there is no specific study available for mastocytosis patients with severe initial SR 233 suggesting the maximum MI to be used in these patients. 234

As first results, it is of note that out of 292 patients only 9.6% of them stopped VIT, 30% delayed their 235 appointment, while 61% respected the original appointment despite the COVID-19 pandemic. Even 236 excluding those patients who did not respect the fixed MI, adherence to VIT remains quite high (61%), 237

considering that the only study available on this topic indicates an adherence rate of 83.7% by the fifth year 238 from the start of VIT in normal situation. 11 239

Our patients are representative of a real-life population with the most common characteristics of HVA 240 patients in terms of demographic, allergy, VIT characteristics, and concomitant conditions (Table I ). In this 241 population, the delay (i.e. extension) from the pre-pandemic scheduled MI to the new post-pandemic MI, 242 due to the COVID-10 pandemic, was often remarkable (up to 16 weeks), causing not only long VIT 243 interruptions (new post-pandemic MI up to 22 weeks) and also abrupt shifts to longer MI (Table II) . 244

However, these extensions were safe and well-tolerated, as only three patients, out of 87, experienced SR 245 (Table III and Table IV ). In our opinion, the main reasons for this achievement could be: 1) the protocol used 246 for extending VIT intervals; 2) the case-by-case approach adopted for the decision making of the time 247 extension, that was weighed against key baseline characteristics; 3) the environment and experience in 248 managing HVA and SR during the treatment. 249

Concerning the adopted protocol, even in case of long delay, it was chosen to resume VIT without starting 250 over with a new build-up phase. In fact, the pre-pandemic maintenance dose was kept and reached in one day in most patients (78%), with consequent avoidance of a new induction phase and therefore a higher 252 patient compliance. The depot extract was used in almost all the patients (95%), probably contributing to 253 the safety 2,12 , even though the low number of aqueous extracts (4 subjects) did not allow a robust risk 254 comparison between them. The facilities of our specialized center for HVA made this protocol feasible in a 255 safe environment, guaranteeing the appropriate management of SE. 256

As for the case-by-case approach, since the pre-pandemic MI already reflected potential patients' risk 257 factors of SE, it was the main factor that affected the decision-making process for the new post-pandemic 258 MI. It is confirmed by the clear significant correlation between pre-and post-pandemic MI, reflecting the 259 caution paid in patients who had already long time intervals between injections before the pandemic. 260

Namely, a long pre-pandemic MI was correlated with shorter delays (p=0.011, Figure 1 ), compared with 261 patients with short pre-pandemic MI. This is consistent with the current recommendations 2,3 that the time 262 interval between VIT injections may be extended, but this should be done progressively over years, after 263 reaching the maintenance dose, especially for VIT efficacy. Moreover, limited data are available on the 264 efficacy and safety of time intervals longer than 12 weeks, 9,10 and fast and long extensions are not 265 recommended in patients who have started VIT recently and/or who have still short intervals (e.g. <4 266 weeks). 267 This is supported by the SR occurred in the patient with the shortest VIT duration (1 month) and pre-268 pandemic MI (2 weeks) of our study population (Table IV , patient #1). A fast and long extension has been 269 required due to pandemic (delay: 10 weeks) in this patient, but it was not tolerated. 270

The severity of the pre-VIT reaction to Hymenoptera stings was another baseline characteristic significantly 271 correlated with the delay (p=0.001, Figure 2A) , and to the subsequent new post-pandemic MI (p=0.047). 272

Indeed, the applied interval extension was shorter, in case of severe baseline reactions, even though 273 currently they are not regarded as risk factors for SR. 2 274

Concerning duration of VIT, the allowed delay between injections was shorter in the patients that had been 275 undergoing VIT for many years, compared to the ones who had started VIT more recently, but this 276 correlation was not linear (p=0.085). The SR experienced by the patient with the shortest VIT duration (patient #1 was the only one undergoing VIT for just 1 month) seems to confirm this risk factor in a non-278 linear way, meaning that there may be risk in delaying too much the VIT administrations shortly after 279 starting VIT (i.e. recent VIT initiation ≤ 1 month), but afterwards the risk does not increase linearly with the 280 VIT duration, in our study. 281

The age was another factor considered for the extension, that was shorter in ≥65-year-old patients, 282 compared with the younger patients (p=0.017, Figure 2B ). However, all the three patients who experienced 283 SRs were quite young (<44 years old). 284

Overall, comorbidities did not play a crucial role to decide the new intervals, because all our patients had 285 well-controlled diseases. However, more cautious was paid in case of multidrug treatments. Consequently, 286 our results do not show a significant relationship between delay and comorbidities, but the delay was 287 significantly shorter for patients undergoing concomitant treatments (p=0.041). 288

On the other hand, the gender, type of HVA, comorbidities, the pre-pandemic threshold of skin reactivity, 289 previous SE related to VIT, diagnosed mastocytosis or potential mastcell disorders (i.e. abnormal REMA 290 score), and high STL were not correlated with the VIT delay, as these factors were not considered limiting 291 factors to extend the intervals, according to our clinical decision making. In fact, the extension of the time 292 intervals (i.e. a mean delay of 8.5 weeks, leading to new post-pandemic MI in the range of 8-22 weeks), in 293 our population, was totally tolerated even in case of comorbidities (including 43% patients with 294 cardiovascular diseases and 5% with mastocytosis), high skin test reactivity (29% patients positive at ≤0.01 295 mcg/ml), history of previous reactions during VIT injections (9%), high STL (10%), and abnormal REMA score 296 (29%). 297

However, notwithstanding that the low number of SRs hinders a proper analysis of risk factors, it is 298 noteworthy that all the three patients with SR share the bee venom allergy, as a statistically significant risk 299 factor (p=0.002). This finding underlines the well-known peculiarity of bee venom. 300

In fact, bee venom allergy has been associated with lower efficacy and higher incidence of SRs to VIT 301 compared with Vespula venoms. 13 The gradual lengthening of the MI up to 12 weeks does not appear to 302 interfere with either efficacy or safety of VIT, 2,3 but it has been documented mainly for Vespid VIT. And only one study showed that further prolonging MI up to 6 months reduces VIT efficacy using bee venom 304 extract while preserving its safety. 9 Thus, more studies are needed in order to safely extend MI or to 305 resume VIT after long delay in bee venom allergic patients, not only in those who have recently started the 306 treatment, but also in long-term VIT. 307

To date, this is the first study providing practical recommendations on the management of VIT during a 308 pandemic, based on the real-life experience with a considerable sample size. The lessons may be applicable 309 in clinical practice not only in case of pandemics, but also whenever any delay in VIT administration is 310 necessary due to other causes (e.g. patients' needs). According to our experience, attention should be paid 311 in extending the time intervals between injections in case of already long MI, severe pre-VIT reaction, older 312 age, and/or multi-drug concomitant treatments. In addition, our results suggest that bee venom allergy and 313 recent initiation of VIT should be considered before extending VIT intervals, as they may be the possible 314 culprits of the SE observed in this study, while a diagnosed or very likely (according to REMA score) mastcell 315 disease does not seem to be a risk factor for prolonging VIT MI. 316

Finally, some limitations must be pointed out in our study. First, the low number of SR is a clinical success 317 and demonstrated the high safety profile of extending the VIT MI in our population, but it makes the 318 association between SE and the identified risk factors (i.e. bee venom) less robust, from a statistical point of 319 view. Second, data come from a single center in Europe where the approach is likely to be different from 320 the USA (i.e. by type of induction phase or extracts used). Therefore, additional studies would be needed to 321 confirm our results. 14.3 ± 3.5 (8-22) 9.0 ± 3.4 (4-16)

pre-pandemic interval ≥8 -<12 w (n=38)

16.8 ± 2.5 (11-22) 8.0 ± 2.7 (3-13)

pre-pandemic interval = 12 w (n=3)

18.3 ± 1.5 (17-20) 6.3 ± 1.5 (5-8) VIT 

Hymenoptera Venom 338 Allergy: Management of Children and Adults in Clinical Practice

EAACI guidelines on allergen 341 immunotherapy: Hymenoptera venom allergy

Treating venom allergy during COVID-19 pandemic

Handling of allergen 347 immunotherapy in the COVID-19 pandemic: An ARIA-EAACI statement

Running a virtual allergy division and training 349 program in the time of COVID-19 pandemic

COVID-19: Pandemic 352 Contingency Planning for the

Validation of the REMA Score for Predicting Mast Cell Clonality and Systemic Mastocytosis in 357 Patients with Systemic Mast Cell Activation Symptoms

Effectiveness of maintenance bee venom immunotherapy 359 administered at 6-month intervals

Continuous immunotherapy for hymenoptera venom allergy 361 using six month intervals

High adherence to 363 hymenoptera venom subcutaneous immunotherapy over a 5-year follow-up: A real-life experience

Purified vs. nonpurified venom immunotherapy

Immunotherapy with honeybee venom and yellow jacket venom is 368 different regarding efficacy and safety